pack-objects: merge read_lock and lock in packing_data struct
Rename the packing_data lock to obd_lock and upgrade it to a recursive
mutex to make it suitable for current read_lock usages. Additionally
remove the superfluous #ifndef NO_PTHREADS guard around mutex
initialization in prepare_packing_data as the mutex functions
themselves are already protected.
Signed-off-by: Patrick Hogg <phogg@novamoon.net> Helped-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
pack-objects: move read mutex to packing_data struct
ac77d0c37 ("pack-objects: shrink size field in struct object_entry",
2018-04-14) added an extra usage of read_lock/read_unlock in the newly
introduced oe_get_size_slow for thread safety in parallel calls to
try_delta(). Unfortunately oe_get_size_slow is also used in serial
code, some of which is called before the first invocation of
ll_find_deltas. As such the read mutex is not guaranteed to be
initialized.
Resolve this by moving the read mutex to packing_data and initializing
it in prepare_packing_data which is initialized in cmd_pack_objects.
Signed-off-by: Patrick Hogg <phogg@novamoon.net> Reviewed-by: Duy Nguyen <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
git-instaweb: add Python builtin http.server support
With this patch it is possible to launch git-instaweb by using
Python http.server CGI handler via `-d python` option.
git-instaweb generates a small wrapper around the http.server
(in GIT_DIR/gitweb/) that address a limitation of the CGI handler
where CGI scripts have to be in a cgi-bin subdirectory and
directory index can't be easily changed. To keep the implementation
small, gitweb is running on url `/cgi-bin/gitweb.cgi` and an automatic
redirection is done when opening `/`.
The generated wrapper is compatible with both Python 2 and 3.
Python is by default installed on most modern Linux distributions
which enables running `git instaweb -d python` without needing
anything else.
Signed-off-by: Arti Zirk <arti.zirk@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
test-date: add a subcommand to measure times in shell scripts
In the next commit, we want to teach Git's test suite to optionally
output test results in JUnit-style .xml files. These files contain
information about the time spent. So we need a way to measure time.
While we could use `date +%s` for that, this will give us only seconds,
i.e. very coarse-grained timings.
GNU `date` supports `date +%s.%N` (i.e. nanosecond-precision output),
but there is no equivalent in BSD `date` (read: on macOS, we would not
be able to obtain precise timings).
So let's introduce `test-tool date getnanos`, with an optional start
time, that outputs preciser values. Note that this might not actually
give us nanosecond precision on some platforms, but it will give us as
precise information as possible, without the portability issues of shell
commands.
Granted, it is a bit pointless to try measuring times accurately in
shell scripts, certainly to nanosecond precision. But it is better than
second-granularity.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
ci: use a junction on Windows instead of a symlink
Symbolic links are still not quite as easy to use on Windows as on Linux
(for example, on versions older than Windows 10, only administrators can
create symlinks, and on Windows 10 you still need to be in developer
mode for regular users to have permission), but NTFS junctions can give
us a way out.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
ci: inherit --jobs via MAKEFLAGS in run-build-and-tests
Let's not decide in the generic ci/ part how many jobs to run in
parallel; different CI configurations would favor a different number of
parallel jobs, and it is easy enough to hand that information down via
the `MAKEFLAGS` variable.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
The upcoming patches will allow building git.git via Azure Pipelines
(i.e. Azure DevOps' Continuous Integration), where variable names and
URLs look a bit different than in Travis CI.
Also, the configurations of the available agents are different. For
example, Travis' and Azure Pipelines' macOS agents are set up
differently, so that on Travis, we have to install the git-lfs and
gettext Homebrew packages, and on Azure Pipelines we do not need to.
Likewise, Azure Pipelines' Ubuntu agents already have asciidoctor
installed.
Finally, on Azure Pipelines the natural way is not to base64-encode tar
files of the trash directories of failed tests, but to publish build
artifacts instead. Therefore, that code to log those base64-encoded tar
files is guarded to be Travis-specific.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
When building a PR, TRAVIS_BRANCH refers to the *target branch*.
Therefore, if a PR targets `master`, and `master` happened to be tagged,
we skipped the build by mistake.
Fix this by using TRAVIS_PULL_REQUEST_BRANCH (i.e. the *source branch*)
when available, falling back to TRAVIS_BRANCH (i.e. for CI builds, also
known as "push builds").
Let's give it a new variable name, too: CI_BRANCH (as it is different
from TRAVIS_BRANCH). This also prepares for the upcoming patches which
will make our ci/* code a bit more independent from Travis and open it
to other CI systems (in particular to Azure Pipelines).
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
implicit interactive rebase: don't run sequence editor
If GIT_SEQUENCE_EDITOR is set then rebase runs it when executing
implicit interactive rebases which are supposed to appear
non-interactive to the user. Fix this by setting GIT_SEQUENCE_EDITOR=:
rather than GIT_EDITOR=:. A couple of tests relied on the old behavior
so they are updated to work with the new regime.
Signed-off-by: Phillip Wood <phillip.wood@dunelm.org.uk> Signed-off-by: Junio C Hamano <gitster@pobox.com>
object_as_type: initialize commit-graph-related fields of 'struct commit'
When the commit graph and generation numbers were introduced in
commits 177722b344 (commit: integrate commit graph with commit
parsing, 2018-04-10) and 83073cc994 (commit: add generation number to
struct commit, 2018-04-25), they tried to make sure that the
corresponding 'graph_pos' and 'generation' fields of 'struct commit'
are initialized conservatively, as if the commit were not included in
the commit-graph file.
Alas, initializing those fields only in alloc_commit_node() missed the
case when an object that happens to be a commit is first looked up via
lookup_unknown_object(), and is then later converted to a 'struct
commit' via the object_as_type() helper function (either calling it
directly, or as part of a subsequent lookup_commit() call).
Consequently, both of those fields incorrectly remain set to zero,
which means e.g. that the commit is present in and is the first entry
of the commit-graph file. This will result in wrong timestamp, parent
and root tree hashes, if such a 'struct commit' instance is later
filled from the commit-graph.
Extract the initialization of 'struct commit's fields from
alloc_commit_node() into a helper function, and call it from
object_as_type() as well, to make sure that it properly initializes
the two commit-graph-related fields, too. With this helper function
it is hopefully less likely that any new fields added to 'struct
commit' in the future would remain uninitialized.
With this change alloc_commit_index() won't have any remaining callers
outside of 'alloc.c', so mark it as static.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff.c: prepare to use parse_options() for parsing
This is a preparation step to start using parse_options() to parse
diff/revision options instead of what we have now. There are a couple
of good things from using parse_options():
- better help usage
- easier to add new options
- better completion support
- help usage generation
- better integration with main command option parser. We can just
concat the main command's option array and diffopt's together and
parse all in one go.
- detect colidding options (e.g. --reverse is used by revision code,
so diff code can't use it as long name for -R)
- consistent syntax, e.g. option that takes mandatory argument will
now accept both "--option=value" and "--option value".
The plan is migrate all diff/rev options to parse_options(). Then we
could get rid of diff_opt_parse() and expose parseopts[] directly to
the caller.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Bitfield addresses cannot be passed around in a pointer. This makes it
hard to use parse-options to set/unset them. Turn this struct to
normal integers. This of course increases the size of this struct
multiple times, but since we only have a handful of diff_options
variables around, memory consumption is not at all a concern.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
parse-options: allow ll_callback with OPTION_CALLBACK
OPTION_CALLBACK is much simpler/safer to use, but parse_opt_cb does
not allow access to parse_opt_ctx_t, which sometimes is useful
(e.g. to obtain the prefix).
Extending parse_opt_cb to take parse_opt_cb could result in a lot of
changes. Instead let's just allow ll_callback to be used with
OPTION_CALLBACK. The user will have to be careful, not to change
anything in ctx, or return wrong result code. But that's the price for
ll_callback.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Give names to these magic negative numbers. Make parse_opt_ll_cb
return an enum to make clear it can actually control parse_options()
with different return values (parse_opt_cb can too, but nobody needs
it).
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
parse-options: disable option abbreviation with PARSE_OPT_KEEP_UNKNOWN
parse-options can unambiguously find an abbreviation only if it sees
all available options. This is usually the case when you use
parse_options(). But there are other callers like blame or shortlog
which uses parse_options_start() in combination with a custom option
parser, like rev-list. parse-options cannot see all options in this
case and will get abbrev detection wrong. Disable it.
t7800 needs update because --symlink no longer expands to --symlinks
and will be passed down to git-diff, which will not recognize it. I
still think this is the correct thing to do. But if --symlink has been
actually used in the wild, we would just add an option alias for it.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
This is to help reimplement diff_opt_parse() using parse_options().
The behavior of parse_options() is changed to be the same as the
other:
- no argv0 in argv[], everything can be processed
- argv[] must not be updated, it's the caller's job to do that
- return the number of arguments processed
- leave all unknown options / non-options alone (this one can already
be achieved with PARSE_OPT_KEEP_UNKNOWN and
PARSE_OPT_STOP_AT_NON_OPTION)
This mode is NOT supposed to stay here for long. It's to help
converting diff/rev option parsing. Once that work is over and we can
just use parse_options() throughout the code base, this will be
deleted.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
strbuf.cocci: suggest strbuf_addbuf() to add one strbuf to an other
The best way to add one strbuf to an other is via:
strbuf_addbuf(&sb, &sb2);
This is a bit more idiomatic and efficient than:
strbuf_addstr(&sb, sb2.buf);
because the size of the second strbuf is known and thus it can spare a
strlen() call, and much more so than:
strbuf_addf(&sb, "%s", sb2.buf);
because it can spare the whole vsnprintf() formatting magic.
Add new semantic patches to 'contrib/coccinelle/strbuf.cocci' to catch
these undesired patterns and to suggest strbuf_addbuf() instead.
Luckily, our codebase is already clean from any such undesired
patterns (but one of the in-flight topics just tried to sneak in such
a strbuf_addf() call).
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Notice that the local side of branch jch starts with "*" instead of
ending with it like the rest. It's not exactly wrong. It just looks
weird.
This patch changes the find-and-replace code a bit to try finding prefix
first before falling back to strstr() which finds a substring from left
to right. Now we have something less OCD
The timestamp we receive is in epoch time, so there's no need for a
timezone parameter to interpret it. The matching show_date() uses "tz"
to show dates in author local time, but relative dates show only the
absolute time difference. The author's location is irrelevant, barring
relativistic effects from using Git close to the speed of light.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
column: drop unused "opts" parameter in item_length()
There are no column options which impact the length computation. In
theory there might be, but this is a file-local function, so it will be
trivial to re-add the parameter should it ever be useful.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
There's no need to pass a header struct to create_bundle(); it writes
the header information directly to a descriptor (and does not report
back details to the caller).
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
apply: drop unused "def" parameter from find_name_gnu()
We use the "def" parameter in find_name_common() for some heuristics,
but they are not necessary with the less-ambiguous GNU format. Let's
drop this unused parameter.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
match-trees: drop unused path parameter from score functions
The scores do not take the particular path into account at
all. It's possible they could, but these are all static file-local
functions. It won't be a big deal to re-add the parameter if they ever
need it.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Currently "--cc --dirstat" will show nothing for a merge. Like
--shortstat and --summary in the previous two patches, it probably makes
sense to treat it like we do --stat, and show a stat against the
first-parent.
This case is less obviously correct than for --shortstat and --summary,
as those are basically variants of --stat themselves. It's possible we
could develop a multi-parent combined dirstat format, in which case we
might regret defining this first-parent behavior. But the same could be
said for --stat, and in the 12+ years of it showing first-parent stats,
nobody has complained.
So showing the first-parent dirstat is at least _useful_, and if we
later develop a clever multi-parent stat format, we'd probably have to
deal with --stat anyway.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Currently "--cc --summary" on a merge shows nothing. Since we show "--cc
--stat" as a stat against the first parent, and because --summary is
typically used in combination with --stat, it makes sense to treat them
both the same way.
Note that we have to tweak t4013's setup a bit to test this case, as the
existing merges do not have any --summary results against their first
parent. But since the merge at the tip of 'master' does add and remove
files with respect to the second parent, we can just make a reversed
doppelganger merge where the parents are swapped.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
The --stat of a combined diff is defined as the first-parent stat,
going all the way back to 965f803c32 (combine-diff: show diffstat with
the first parent., 2006-04-17).
Naturally, we gave --numstat the same treatment in 74e2abe5b7 (diff
--numstat, 2006-10-12).
But --shortstat, which is really just the final line of --stat, does
nothing, which produces confusing results:
$ git show --oneline --stat eab7584e37 eab7584e37 Merge branch 'en/show-ref-doc-fix'
There are several conditionals in the combine diff code that check if
we're doing --stat or --numstat output. Since these must all remain in
sync, let's pull them out into a separate bit-mask.
Arguably this could go into diff.h along with the other DIFF_FORMAT
macros, but it's not clear that the definition of "which formats are
stat" is universal (e.g., does --dirstat count? --summary?). So let's
keep this local to combine-diff.c, where the meaning is more clearly
"stat formats that combine-diff supports".
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
There's an odd bug when "log --color-moved" is used with the combination
of "--cc --stat -p": the stat for merge commits is erroneously shown
with the diff of the _next_ commit.
The included test demonstrates the issue. Our history looks something
like this:
A-B-M--D
\ /
C
When we run "git log --cc --stat -p --color-moved" starting at D, we get
this sequence of events:
1. The diff for D is using -p, so diff_flush() calls into
diff_flush_patch_all_file_pairs(). There we see that o->color_moved
is in effect, so we point o->emitted_symbols to a static local
struct, causing diff_flush_patch() to queue the symbols instead of
actually writing them out.
We then do our move detection, emit the symbols, and clear the
struct. But we leave o->emitted_symbols pointing to our struct.
2. Next we compute the diff for M. This is a merge, so we use the
combined diff code. In find_paths_generic(), we compute the
pairwise diff between each commit and its parent. Normally this is
done with DIFF_FORMAT_NO_OUTPUT, since we're just looking for
intersecting paths. But since "--stat --cc" shows the first-parent
stat, and since we're computing that diff anyway, we enable
DIFF_FORMAT_DIFFSTAT for the first parent. This outputs the stat
information immediately, saving us from running a separate
first-parent diff later.
But where does that output go? Normally it goes directly to stdout,
but because o->emitted_symbols is set, we queue it. As a result, we
don't actually print the diffstat for the merge commit (yet), which
is wrong.
3. Next we compute the diff for C. We're actually showing a patch
again, so we end up in diff_flush_patch_all_file_pairs(), but this
time we have the queued stat from step 2 waiting in our struct.
We add new elements to it for C's diff, and then flush the whole
thing. And we see the diffstat from M as part of C's diff, which is
wrong.
So triggering the bug really does require the combination of all of
those options.
To fix it, we can simply restore o->emitted_symbols to NULL after
flushing it, so that it does not affect anything outside of
diff_flush_patch_all_file_pairs(). This intuitively makes sense, since
nobody outside of that function is going to bother flushing it, so we
would not want them to write to it either.
In fact, we could take this a step further and turn the local "esm"
struct into a non-static variable that goes away after the function
ends. However, since it contains a dynamically sized array, we benefit
from amortizing the cost of allocations over many calls. So we'll leave
it as static to retain that benefit.
But let's push the zero-ing of esm.nr into the conditional for "if
(o->emitted_symbols)" to make it clear that we do not expect esm to hold
any values if we did not just try to use it. With the code as it is
written now, if we did encounter such a case (which I think would be a
bug), we'd silently leak those values without even bothering to display
them. With this change, we'd at least eventually show them, and somebody
would notice.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
This set of tests was added by 4434e6ba6c (tests: check --[short]stat
output after chmod, 2012-05-01), and is primarily about the handling of
binary versus text files.
Later, 74faaa16f0 (Fix "git diff --stat" for interesting - but empty -
file changes, 2012-10-17) changed the stat output so that the empty text
file is mentioned rather than omitted. That commit just comments out
these tests. There's no discussion in the commit message, but the
original email[1] says:
NOTE! This does break two of our tests, so we clearly did this on
purpose, or at least tested for it. I just uncommented the subtests
that this makes irrelevant, and changed the output of another one.
I don't think they're irrelevant, though. We should be testing this
"mode change only" case and making sure that it has the post-74faaa16f0
behavior. So this commit brings back those tests, with the current
expected output.
checkout: fix regression in checkout -b on intitial checkout
When doing a 'checkout -b' do a full checkout including updating the working
tree when doing the initial checkout. As the new test involves an filesystem
access, do it later in the sequence to give chance to other cheaper tests to
leave early. This fixes the regression in behavior caused by fa655d8411
(checkout: optimize "git checkout -b <new_branch>", 2018-08-16).
Signed-off-by: Ben Peart <benpeart@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
checkout: add test demonstrating regression with checkout -b on initial commit
Commit fa655d8411 (checkout: optimize "git checkout -b <new_branch>",
2018-08-16) introduced an unintentional change in behavior for 'checkout -b'
after doing 'clone --no-checkout'. Add a test to demonstrate the changed
behavior to be used in a later patch to verify the fix.
Signed-off-by: Ben Peart <benpeart@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
commit-graph write: emit a percentage for all progress
Follow-up 01ca387774 ("commit-graph: split up close_reachable()
progress output", 2018-11-19) by making the progress bars in
close_reachable() report a completion percentage. This fixes the last
occurrence where in the commit graph writing where we didn't report
that.
The change in 01ca387774 split up the 1x progress bar in
close_reachable() into 3x, but left them as dumb counters without a
percentage completion. Fixing that is easy, and the only reason it
wasn't done already is because that commit was rushed in during the
v2.20.0 RC period to fix the unrelated issue of over-reporting commit
numbers. See [1] and follow-ups for ML activity at the time and [2]
for an alternative approach where the progress bars weren't split up.
Now for e.g. linux.git we'll emit:
$ ~/g/git/git --exec-path=$HOME/g/git commit-graph write
Finding commits for commit graph among packed objects: 100% (6529159/6529159), done.
Expanding reachable commits in commit graph: 100% (815990/815980), done.
Computing commit graph generation numbers: 100% (815983/815983), done.
Writing out commit graph in 4 passes: 100% (3263932/3263932), done.
Add progress output to sections of code between "Annotating[...]" and
"Computing[...]generation numbers". This can collectively take 5-10
seconds on a large enough repository.
On a test repository with I have with ~7 million commits and ~50
million objects we'll now emit:
$ ~/g/git/git --exec-path=$HOME/g/git commit-graph write
Finding commits for commit graph among packed objects: 100% (124763727/124763727), done.
Loading known commits in commit graph: 100% (18989461/18989461), done.
Expanding reachable commits in commit graph: 100% (18989507/18989461), done.
Clearing commit marks in commit graph: 100% (18989507/18989507), done.
Counting distinct commits in commit graph: 100% (18989507/18989507), done.
Finding extra edges in commit graph: 100% (18989507/18989507), done.
Computing commit graph generation numbers: 100% (7250302/7250302), done.
Writing out commit graph in 4 passes: 100% (29001208/29001208), done.
Whereas on a medium-sized repository such as linux.git these new
progress bars won't have time to kick in and as before and we'll still
emit output like:
$ ~/g/git/git --exec-path=$HOME/g/git commit-graph write
Finding commits for commit graph among packed objects: 100% (6529159/6529159), done.
Expanding reachable commits in commit graph: 815990, done.
Computing commit graph generation numbers: 100% (815983/815983), done.
Writing out commit graph in 4 passes: 100% (3263932/3263932), done.
The "Counting distinct commits in commit graph" phase will spend most
of its time paused at "0/*" as we QSORT(...) the list. That's not
optimal, but at least we don't seem to be stalling anymore most of the
time.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
commit-graph write: remove empty line for readability
Remove the empty line between a QSORT(...) and the subsequent oideq()
for-loop. This makes it clearer that the QSORT(...) is being done so
that we can run the oideq() loop on adjacent OIDs. Amends code added
in 08fd81c9b6 ("commit-graph: implement write_commit_graph()",
2018-04-02).
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
commit-graph write: add more descriptive progress output
Make the progress output shown when we're searching for commits to
include in the graph more descriptive. This amends code I added in 7b0f229222 ("commit-graph write: add progress output", 2018-09-17).
Now, on linux.git, we'll emit this sort of output in the various modes
we support:
$ git commit-graph write
Finding commits for commit graph among packed objects: 100% (6529159/6529159), done.
[...]
# Actually we don't emit this since this takes almost no time at
# all. But if we did (s/_delayed//) we'd show:
$ git for-each-ref --format='%(objectname)' | git commit-graph write --stdin-commits
Finding commits for commit graph from 630 refs: 100% (630/630), done.
[...]
$ (cd .git/objects/pack/ && ls *idx) | git commit-graph write --stdin-pack
Finding commits for commit graph in 3 packs: 6529159, done.
[...]
The middle on of those is going to be the output users might see in
practice, since it'll be emitted when they get the commit graph via
gc.writeCommitGraph=true. But as noted above you need a really large
number of refs for this message to show. It'll show up on a test
repository I have with ~165k refs:
Finding commits for commit graph from 165203 refs: 100% (165203/165203), done.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
commit-graph write: show progress for object search
Show the percentage progress for the "Finding commits for commit
graph" phase for the common case where we're operating on all packs in
the repository, as "commit-graph write" or "gc" will do.
Before we'd emit on e.g. linux.git with "commit-graph write":
Finding commits for commit graph: 6529159, done.
[...]
And now:
Finding commits for commit graph: 100% (6529159/6529159), done.
[...]
Since the commit graph only includes those commits that are packed
(via for_each_packed_object(...)) the approximate_object_count()
returns the actual number of objects we're going to process.
Still, it is possible due to a race with "gc" or another process
maintaining packs that the number of objects we're going to process is
lower than what approximate_object_count() reported. In that case we
don't want to stop the progress bar short of 100%. So let's make sure
it snaps to 100% at the end.
The inverse case is also possible and more likely. I.e. that a new
pack has been added between approximate_object_count() and
for_each_packed_object(). In that case the percentage will go beyond
100%, and we'll do nothing to snap it back to 100% at the end.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
commit-graph write: more descriptive "writing out" output
Make the "Writing out" part of the progress output more
descriptive. Depending on the shape of the graph we either make 3 or 4
passes over it.
Let's present this information to the user in case they're wondering
what this number, which is much larger than their number of commits,
has to do with writing out the commit graph. Now e.g. on linux.git we
emit:
$ ~/g/git/git --exec-path=$HOME/g/git -C ~/g/linux commit-graph write
Finding commits for commit graph: 6529159, done.
Expanding reachable commits in commit graph: 815990, done.
Computing commit graph generation numbers: 100% (815983/815983), done.
Writing out commit graph in 4 passes: 100% (3263932/3263932), done.
A note on i18n: Why are we using the Q_() function and passing a
number & English text for a singular which'll never be used? Because
the plural rules of translated languages may not match those of
English, and to use the plural function we need to use this format.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add progress output to be shown when we're writing out the
commit-graph, this adds to the output already added in 7b0f229222
("commit-graph write: add progress output", 2018-09-17).
As noted in that commit most of the progress output isn't displayed on
small repositories, but before this change we'd noticeably hang for
2-3 seconds at the end on medium sized repositories such as linux.git.
Now we'll instead show output like this, and reduce the
human-observable times at which we're not producing progress output:
This "Writing out" number is 3x or 4x the number of commits, depending
on the graph we're processing. A later change will make this explicit
to the user.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
The optional 'Extra Edge List' chunk of the commit graph file stores
parent information for commits with more than two parents. Since the
chunk is optional, write_commit_graph() looks through all commits to
find those with more than two parents, and then writes the commit
graph file header accordingly, i.e. if there are no such commits, then
there won't be a 'Extra Edge List' chunk written, only the three
mandatory chunks.
However, when it later comes to writing actual chunk data,
write_commit_graph() unconditionally invokes
write_graph_chunk_extra_edges(), even when it was decided earlier that
that chunk won't be written. Strictly speaking there is no bug here,
because write_graph_chunk_extra_edges() won't write anything if it
doesn't find any commits with more than two parents, but then it
unnecessarily and in vain looks through all commits once again in
search for such commits.
Don't call write_graph_chunk_extra_edges() when that chunk won't be
written to spare an unnecessary iteration over all commits.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Display date and time information in a format similar to how people
write dates in other contexts. If the year isn't specified then, the
reader infers the date is given is in the current year.
By not displaying the redundant information, the reader concentrates
on the information that is different. The patch reports relative dates
based on information inferred from the date on the machine running the
git command at the time the command is executed.
While the format is more useful to humans by dropping inferred
information, there is nothing that makes it actually human. If the
'relative' date format wasn't already implemented then using
'relative' would have been appropriate.
Signed-off-by: Stephen P. Smith <ischis2@cox.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
In addition to adding the 'human' format, the patch added the auto
keyword which could be used in the config file as an alternate way to
specify the human format. Removing 'auto' cleans up the 'human'
format interface.
Added the ability to specify mode 'foo' if the pager is being used by
using auto:foo syntax. Therefore, 'auto:human' date mode defaults to
human if we're using the pager. So you can do
git config --add log.date auto:human
and your "git log" commands will show the human-legible format unless
you're scripting things.
Signed-off-by: Stephen P. Smith <ischis2@cox.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since 60a12722ac (attr: remove maybe-real, maybe-macro from git_attr,
2017-01-27), we will always mark an attribute macro (e.g., "binary")
that is specifically queried for as "unspecified", even though listing
_all_ attributes would display it at set. E.g.:
$ echo "* binary" >.gitattributes
$ git check-attr -a file
file: binary: set
file: diff: unset
file: merge: unset
file: text: unset
The problem stems from an incorrect conversion of the optimization from 06a604e670 (attr: avoid heavy work when we know the specified attr is
not defined, 2014-12-28). There we tried in collect_some_attrs() to
avoid even looking at the attr_stack when the user has asked for "foo"
and we know that "foo" did not ever appear in any .gitattributes file.
It used a flag "maybe_real" in each attribute struct, where "real" meant
that the attribute appeared in an actual file (we have to make this
distinction because we also create an attribute struct for any names
that are being queried). But as explained in that commit message, the
meaning of "real" was tangled with some special cases around macros.
When 60a12722ac later refactored the macro code, it dropped maybe_real
entirely. This missed the fact that "maybe_real" could be unset for two
reasons: because of a macro, or because it was never found during
parsing. This had two results:
- the optimization in collect_some_attrs() ceased doing anything
meaningful, since it no longer kept track of "was it found during
parsing"
- worse, it actually kicked in when the caller _did_ ask about a macro
by name, causing us to mark it as unspecified
It should be possible to salvage this optimization, but let's start with
just removing the remnants. It hasn't been doing anything (except
creating bugs) since 60a12722ac, and nobody seems to have noticed the
performance regression. It's more important to fix the correctness
problem clearly first.
I've added two tests here. The second one actually shows off the bug.
The test of "check-attr -a" is not strictly necessary, but we currently
do not test attribute macros much, and the builtin "binary" not at all.
So this increases our general test coverage, as well as making sure we
didn't mess up this related case.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
In 8abfdf44c882 (tests: explicitly use `git.exe` on Windows,
2018-11-14), we made sure to use the `.exe` file extension when
using an absolute path to `git.exe`, to avoid getting confused with a
file or directory in the same place that lacks said file extension.
For the same reason, we need to handle test-tool.exe the same way.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
commit-graph: rename "large edges" to "extra edges"
The optional 'Large Edge List' chunk of the commit graph file stores
parent information for commits with more than two parents, and the
names of most of the macros, variables, struct fields, and functions
related to this chunk contain the term "large edges", e.g.
write_graph_chunk_large_edges(). However, it's not a really great
term, as the edges to the second and subsequent parents stored in this
chunk are not any larger than the edges to the first and second
parents stored in the "main" 'Commit Data' chunk. It's the number of
edges, IOW number of parents, that is larger compared to non-merge and
"regular" two-parent merge commits. And indeed, two functions in
'commit-graph.c' have a local variable called 'num_extra_edges' that
refer to the same thing, and this "extra edges" term is much better at
describing these edges.
So let's rename all these references to "large edges" in macro,
variable, function, etc. names to "extra edges". There is a
GRAPH_OCTOPUS_EDGES_NEEDED macro as well; for the sake of consistency
rename it to GRAPH_EXTRA_EDGES_NEEDED.
We can do so safely without causing any incompatibility issues,
because the term "large edges" doesn't come up in the file format
itself in any form (the chunk's magic is {'E', 'D', 'G', 'E'}, there
is no 'L' in there), but only in the specification text. The string
"large edges", however, does come up in the output of 'git
commit-graph read' and in tests looking at its input, but that command
is explicitly documented as debugging aid, so we can change its output
and the affected tests safely.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
commit-graph write: use pack order when finding commits
Slightly optimize the "commit-graph write" step by using
FOR_EACH_OBJECT_PACK_ORDER with for_each_object_in_pack(). See commit
[1] and [2] for the facility and a similar optimization for "cat-file".
Than the same thing with the "cat" omitted. This is as expected, since
we're iterating in pack order and the "cat" is extra work.
Before this change the opposite was true of "commit-graph write". We
were 6% faster if we first ran "cat" to efficiently populate the FS
cache for our sole big pack on linux.git, than if we had populated it
via for_each_object_in_pack(). Now we're 3% faster without the "cat"
instead.
My tests were done on an unloaded Linux 3.10 system with 10 runs for
each. Derrick Stolee did his own tests on Windows[3] showing a 2%
improvement with a high degree of accuracy.
1. 736eb88fdc ("for_each_packed_object: support iterating in
pack-order", 2018-08-10)
2. 0750bb5b51 ("cat-file: support "unordered" output for
--batch-all-objects", 2018-08-10)
If `git commit-tree HEAD^{tree}` fails on us and produces no output on
stdout, we will substitute that empty string and execute `git tag
ninth-unsigned`, i.e., we will tag HEAD rather than a newly created
object. But we are lucky: we have a signature on HEAD, so we should
eventually fail the next test, where we verify that "ninth-unsigned" is
indeed unsigned.
We have a similar problem a few lines later. If `git commit-tree -S`
fails with no output, we will happily tag HEAD as "tenth-signed". Here,
we are not so lucky. The tag ends up on the same commit as
"eighth-signed-alt", and that's a signed commit, so t7510-signed-commit
will pass, despite `git commit-tree -S` failing.
Make these `git commit-tree` invocations a direct part of the &&-chain,
so that we can rely less on luck and set a better example for future
tests modeled after this one. Fix a 9/10 copy/paste error while at it.
Signed-off-by: Martin Ågren <martin.agren@gmail.com> Signed-off-by: Brandon Richardson <brandon1024.br@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
git-submodule: abort if core.worktree could not be set correctly
74d4731da1f (submodule--helper: replace connect-gitdir-workingtree by
ensure-core-worktree, 2018-08-13) forgot to exit the submodule operation
if the helper could not ensure that core.worktree is set correctly.
Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Do not print 'dangling' for cat-file in case of ambiguity
The return values -1 and -2 from get_oid could mean two different
things, depending on whether they were from an enum returned by
get_tree_entry_follow_symlinks, or from a different code path. This
caused 'dangling' to be printed from a git cat-file in the case of an
ambiguous (-2) result.
Unify the results of get_oid* and get_tree_entry_follow_symlinks to be
one common type, with unambiguous values.
Signed-off-by: David Turner <novalis@novalis.org> Reported-by: Eric Wong <e@80x24.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git gc" and "git repack" did not close the open packfiles that
they found unneeded before removing them, which didn't work on a
platform incapable of removing an open file. This has been
corrected.
* js/gc-repack-close-before-remove:
gc/repack: release packs when needed
The "--format=<placeholder>" option of for-each-ref, branch and tag
learned to show a few more traits of objects that can be learned by
the object_info API.
* ot/ref-filter-object-info:
ref-filter: give uintmax_t to format with %PRIuMAX
ref-filter: add docs for new options
ref-filter: add tests for deltabase
ref-filter: add deltabase option
ref-filter: add tests for objectsize:disk
ref-filter: add check for negative file size
ref-filter: add objectsize:disk option
Flaky tests can now be repeatedly run under load with the
"--stress" option.
* sg/stress-test:
test-lib: add the '--stress' option to run a test repeatedly under load
test-lib-functions: introduce the 'test_set_port' helper function
test-lib: set $TRASH_DIRECTORY earlier
test-lib: consolidate naming of test-results paths
test-lib: parse command line options earlier
test-lib: parse options in a for loop to keep $@ intact
test-lib: extract Bash version check for '-x' tracing
test-lib: translate SIGTERM and SIGHUP to an exit
The loose object cache used to optimize existence look-up has been
updated.
* rs/loose-object-cache-perffix:
object-store: retire odb_load_loose_cache()
object-store: use one oid_array per subdirectory for loose cache
object-store: factor out odb_clear_loose_cache()
object-store: factor out odb_loose_cache()
Earlier we added "-Wformat-security" to developer builds, assuming
that "-Wall" (which includes "-Wformat" which in turn is required
to use "-Wformat-security") is always in effect. This is not true
when config.mak.autogen is in use, unfortunately. This has been
fixed by unconditionally adding "-Wall" to developer builds.
* jk/dev-build-format-security:
config.mak.dev: add -Wall, primarily for -Wformat, to help autoconf users
"git cherry-pick -m1" was forbidden when picking a non-merge
commit, even though there _is_ parent number 1 for such a commit.
This was done to avoid mistakes back when "cherry-pick" was about
picking a single commit, but is no longer useful with "cherry-pick"
that can pick a range of commits. Now the "-m$num" option is
allowed when picking any commit, as long as $num names an existing
parent of the commit.
Technically this is a backward incompatible change; hopefully
nobody is relying on the error-checking behaviour.
* so/cherry-pick-always-allow-m1:
t3506: validate '-m 1 -ff' is now accepted for non-merge commits
t3502: validate '-m 1' argument is now accepted for non-merge commits
cherry-pick: do not error on non-merge commits when '-m 1' is specified
t3510: stop using '-m 1' to force failure mid-sequence of cherry-picks
"git worktree remove" and "git worktree move" refused to work when
there is a submodule involved. This has been loosened to ignore
uninitialized submodules.
* nd/worktree-remove-with-uninitialized-submodules:
worktree: allow to (re)move worktrees with uninitialized submodules
The test suite tried to see if it is run under bash, but the check
itself failed under some other implementations of shell (notably
under NetBSD). This has been corrected.
* sg/test-bash-version-fix:
test-lib: check Bash version for '-x' without using shell arrays
* rb/hpe:
compat/regex/regcomp.c: define intptr_t and uintptr_t on NonStop
git-compat-util.h: add FLOSS headers for HPE NonStop
config.mak.uname: support for modern HPE NonStop config.
transport-helper: drop read/write errno checks
transport-helper: use xread instead of read
With zsh, "git cmd path<TAB>" was completed to "git cmd path name"
when the completed path has a special character like SP in it,
without any attempt to keep "path name" a single filename. This
has been fixed to complete it to "git cmd path\ name" just like
Bash completion does.
* cy/zsh-completion-SP-in-path:
completion: treat results of git ls-tree as file paths
zsh: complete unquoted paths with spaces correctly
The core.worktree setting in a submodule repository should not be
pointing at a directory when the submodule loses its working tree
(e.g. getting deinit'ed), but the code did not properly maintain
this invariant.
* sb/submodule-unset-core-worktree-when-worktree-is-lost:
submodule deinit: unset core.worktree
submodule--helper: fix BUG message in ensure_core_worktree
submodule: unset core.worktree if no working tree is present
submodule update: add regression test with old style setups
The MSYS2 runtime does its best to emulate the command-line wildcard
expansion and de-quoting which would be performed by the calling Unix
shell on Unix systems.
Those Unix shell quoting rules differ from the quoting rules applying to
Windows' cmd and Powershell, making it a little awkward to quote
command-line parameters properly when spawning other processes.
In particular, git.exe passes arguments to subprocesses that are *not*
intended to be interpreted as wildcards, and if they contain
backslashes, those are not to be interpreted as escape characters, e.g.
when passing Windows paths.
Note: this is only a problem when calling MSYS2 executables, not when
calling MINGW executables such as git.exe. However, we do call MSYS2
executables frequently, most notably when setting the use_shell flag in
the child_process structure.
There is no elegant way to determine whether the .exe file to be
executed is an MSYS2 program or a MINGW one. But since the use case of
passing a command line through the shell is so prevalent, we need to
work around this issue at least when executing sh.exe.
Let's introduce an ugly, hard-coded test whether argv[0] is "sh", and
whether it refers to the MSYS2 Bash, to determine whether we need to
quote the arguments differently than usual.
That still does not fix the issue completely, but at least it is
something.
Incidentally, this also fixes the problem where `git clone \\server\repo`
failed due to incorrect handling of the backslashes when handing the path
to the git-upload-pack process.
Further, we need to take care to quote not only whitespace and
backslashes, but also curly brackets. As aliases frequently go through
the MSYS2 Bash, and as aliases frequently get parameters such as
HEAD@{yesterday}, this is really important. As an early version of this
patch broke this, let's make sure that this does not regress by adding a
test case for that.
Helped-by: Kim Gybels <kgybels@infogroep.be> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
mingw (t5580): document bug when cloning from backslashed UNC paths
Due to a quirk in Git's method to spawn git-upload-pack, there is a
problem when passing paths with backslashes in them: Git will force the
command-line through the shell, which has different quoting semantics in
Git for Windows (being an MSYS2 program) than regular Win32 executables
such as git.exe itself.
The symptom is that the first of the two backslashes in UNC paths of the
form \\myserver\folder\repository.git is *stripped off*.
Document this bug by introducing a test case.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
ls-refs: filter refs using namespace-stripped name
If a user fetches refs/heads/master from a repo with namespace "ns", the
remote is expected to (1) not send the real refs/heads/master, and (2)
send refs/namespaces/ns/refs/heads/master with the name
refs/heads/master. (1) indeed happens now, but not (2) - Git only sends
refs that have the user-given prefix, but it checks them against the
full name of the ref (the one starting with refs/namespaces), and not
the namespace-stripped one.
This is demonstrated by the patch in the test. Currently, it results in
"fatal: couldn't find remote ref refs/heads/master" despite both
unnamespaced and namespaced master being present. With the code change,
it produces the expected result.
Check the ref prefixes against the namespace-stripped name.
This bug was discovered through applying patches [1] that override
protocol.version to 2 in repositories when running tests, allowing us to
notice differences in behavior across different protocol versions.
This adds --date=human, which skips the timezone if it matches the
current time-zone, and doesn't print the whole date if that matches (ie
skip printing year for dates that are "this year", but also skip the
whole date itself if it's in the last few days and we can just say what
weekday it was).
For really recent dates (same day), use the relative date stamp, while
for old dates (year doesn't match), don't bother with time and timezone.
Also add 'auto' date mode, which defaults to human if we're using the
pager. So you can do
git config --add log.date auto
and your "git log" commands will show the human-legible format unless
you're scripting things.
Note that this time format still shows the timezone for recent enough
events (but not so recent that they show up as relative dates). You can
combine it with the "-local" suffix to never show timezones for an even
more simplified view.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Stephen P. Smith <ischis2@cox.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
While the scripted `git rebase` still has to rely on the
`git-rebase--am.sh` script to implement the glue between the `rebase`
and the `am` commands, we can go a more direct route in the built-in
rebase and avoid using a shell script altogether.
This patch represents a straight-forward port of `git-rebase--am.sh` to
C, along with the glue code to call it directly from within
`builtin/rebase.c`.
This reduces the chances of Git for Windows running into trouble due to
problems with the POSIX emulation layer (known as "MSYS2 runtime",
itself a derivative of the Cygwin runtime): when no shell script is
called, the POSIX emulation layer is avoided altogether.
Note: we pass an empty action to `reset_head()` here when moving back to
the original branch, as no other action is applicable, really. This
parameter is used to initialize `unpack_trees()`' messages.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
rebase: avoid double reflog entry when switching branches
When switching a branch *and* updating said branch to a different
revision, let's avoid a double entry in HEAD's reflog by first updating
the branch and then adjusting the symbolic ref HEAD.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Over the next commits, we want to make use of it in `run_am()` (i.e.
running the `--am` backend directly, without detouring to Unix shell
script code) which in turn will be called from `run_specific_rebase()`.
So let's move it before that latter function.
This commit is best viewed using --color-moved.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>